To track a maneuvering hypersonic glide target, an adaptive method for state estimation is proposed with model uncertainties in this paper. Various maneuver modes lead to considerable motion model uncertainty. In the method, unknown aerodynamic accelerations are modeled as the Singer model with a small maneuver frequency. When the real motion mode deviates from the default model, the maneuver frequency is adaptively enlarged by a novel technique derived from the orthogonal principle to reduce the model error. Simultaneously, the Huber technique is adopted to address measurement model uncertainty. Two strategies are specially formulated to alleviate possible misjudgments that a type of model error activates the non-corresponding technique. The adaptive state estimation is realized under the framework of unscented Kalman filter. Through tracking different flight trajectories, simulation results demonstrate that the proposed method has stronger robustness and higher estimation accuracy than conventional methods in the presence of model uncertainties and the computation burden is significantly less than the multiple-model method.

针对机动高超声速滑翔目标的跟踪，提出了一种模型不确定的自适应状态估计方法。各种机动模式导致相当大的运动模型不确定性。该方法将未知气动加速度建模为机动频率小的Singer模型。当实际运动模式偏离默认模型时，通过正交原理导出的新技术自适应地增大机动频率以减少模型误差。同时，采用Huber技术解决测量模型的不确定性问题。专门制定了两种策略，以减轻一种模型错误激活非对应技术可能产生的误判。自适应状态估计是在无迹卡尔曼滤波器的框架下实现的。